Merge pull request #199 from NodeFactoryIo/Release-0.5.4

Release 0.5.4

Author: mpetrunic

.version

@@ -1 +1 @@
-version=0.5.3
+version=0.5.4

CHANGELOG.md

@@ -1,12 +1,13 @@
 # Changelog
 
-## [unreleased]((https://github.com/NodeFactoryIo/vedran/tree/HEAD))
-[Full Changelog](https://github.com/NodeFactoryIo/vedran/compare/v0.5.3...HEAD)
+## [v0.5.4]((https://github.com/NodeFactoryIo/vedran/tree/v0.5.4))
+[Full Changelog](https://github.com/NodeFactoryIo/vedran/compare/v0.5.3...v0.5.4)
 
 ### Added
 
 ### Fix
 - Fix tunnel tcp connections not closing after requests finish [\#197](https://github.com/NodeFactoryIo/vedran/pull/197) ([mpetrun5](https://github.com/mpetrun5))
+- Ping handling upgrade [\#198](https://github.com/NodeFactoryIo/vedran/pull/198) ([MakMuftic](https://github.com/MakMuftic))
 
 ### Changed
 

docker-compose.yml

@@ -26,11 +26,11 @@ services:
     container_name: "vedran"
 
   vedran-daemon:
-    image: nodefactory/vedran-daemon:v0.3.2
+    image: nodefactory/vedran-daemon:v0.3.4
     depends_on:
       - vedran
       - polkadot
-    command: --id test-id --lb http://vedran:4000 --node-rpc http://polkadot:9933 --node-ws http://polkadot:9944 --node-metrics http://polkadot:9615 --payout-address 1Z4GTfUN2QHkSeHqdBUDawgbEWzqtfRG8ouJZ26z3cm7ePN --log-level info
+    command: --id test-id --lb http://vedran:4000 --node-rpc http://polkadot:9933 --node-ws ws://polkadot:9944 --node-metrics http://polkadot:9615 --payout-address 1Z4GTfUN2QHkSeHqdBUDawgbEWzqtfRG8ouJZ26z3cm7ePN --log-level info
 
   prometheus:
     image: prom/prometheus

internal/controllers/ping.go

@@ -10,7 +10,7 @@ import (
 	log "github.com/sirupsen/logrus"
 )
 
-const pingOffset = 5
+const pingOffset = 8
 
 func (c ApiController) PingHandler(w http.ResponseWriter, r *http.Request) {
 	request := r.Context().Value(auth.RequestContextKey).(*auth.RequestContext)
@@ -20,18 +20,24 @@ func (c ApiController) PingHandler(w http.ResponseWriter, r *http.Request) {
 		log.Errorf("Unable to calculate node downtime, error: %v", err)
 	}
 
-	if math.Abs(downtimeDuration.Seconds()) > (stats.PingIntervalInSeconds + pingOffset) {
-		downtime := models.Downtime{
-			Start:  lastPingTime,
-			End:    request.Timestamp,
-			NodeId: request.NodeId,
+	// if two pings come one after another (in 2 second interval)
+	// this means that one ping stuck in network and
+	// there is no need to write multiple downtimes
+	if math.Abs(request.Timestamp.Sub(lastPingTime).Seconds()) > 2 {
+		// check if there were downtime
+		if math.Abs(downtimeDuration.Seconds()) > (stats.PingIntervalInSeconds + pingOffset) {
+			downtime := models.Downtime{
+				Start:  lastPingTime,
+				End:    request.Timestamp,
+				NodeId: request.NodeId,
+			}
+			err = c.repositories.DowntimeRepo.Save(&downtime)
+			if err != nil {
+				log.Errorf("Unable to save node downtime, error: %v", err)
+			}
+
+			log.Debugf("Saved node %s downtime of: %f", request.NodeId, math.Abs(downtimeDuration.Seconds()))
 		}
-		err = c.repositories.DowntimeRepo.Save(&downtime)
-		if err != nil {
-			log.Errorf("Unable to save node downtime, error: %v", err)
-		}
-
-		log.Debugf("Saved node %s downtime of: %f", request.NodeId, math.Abs(downtimeDuration.Seconds()))
 	}
 
 	// save ping to database

internal/controllers/ping_test.go

@@ -27,6 +27,8 @@ func TestApiController_PingHandler(t *testing.T) {
 		downtimeSaveErr       error
 		calculateDowntimeErr  error
 		downtimeDuration      time.Duration
+		requestTimestamp time.Time
+		lastPingTimestamp time.Time
 	}{
 		{
 			name:                  "Returns 200 if downtime calculation fails",
@@ -37,26 +39,32 @@ func TestApiController_PingHandler(t *testing.T) {
 			downtimeSaveCallCount: 0,
 			downtimeDuration:      time.Duration(0),
 			calculateDowntimeErr:  fmt.Errorf("ERROR"),
+			requestTimestamp: time.Now(),
+			lastPingTimestamp: time.Now().Add(-5 * time.Second),
 		},
 		{
-			name:                  "Returns 200 if donwtime save fails",
+			name:                  "Returns 200 if downtime save fails",
 			statusCode:            200,
 			pingSaveCallCount:     1,
 			pingSaveErr:           nil,
 			downtimeSaveErr:       fmt.Errorf("ERROR"),
 			downtimeSaveCallCount: 1,
-			downtimeDuration:      time.Duration(time.Second * 11),
+			downtimeDuration:      time.Duration(time.Second * 19),
 			calculateDowntimeErr:  nil,
+			requestTimestamp: time.Now(),
+			lastPingTimestamp: time.Now().Add(-19 * time.Second),
 		},
 		{
-			name:                  "Saves downtime if downtime duration more than 5 seconds",
+			name:                  "Saves downtime if downtime duration more than 18 seconds",
 			statusCode:            200,
 			pingSaveCallCount:     1,
 			pingSaveErr:           nil,
 			downtimeSaveErr:       nil,
 			downtimeSaveCallCount: 1,
-			downtimeDuration:      time.Duration(time.Second * 11),
+			downtimeDuration:      time.Duration(time.Second * 19),
 			calculateDowntimeErr:  nil,
+			requestTimestamp: time.Now(),
+			lastPingTimestamp: time.Now().Add(-19 * time.Second),
 		},
 		{
 			name:                  "Returns 500 if saving ping fails",
@@ -67,6 +75,8 @@ func TestApiController_PingHandler(t *testing.T) {
 			downtimeSaveCallCount: 0,
 			downtimeDuration:      time.Duration(time.Second * 8),
 			calculateDowntimeErr:  nil,
+			requestTimestamp: time.Now(),
+			lastPingTimestamp: time.Now().Add(-5 * time.Second),
 		},
 		{
 			name:                  "Returns 200 and does not save downtime if downtime duration less than 5 + 5 seconds",
@@ -77,13 +87,14 @@ func TestApiController_PingHandler(t *testing.T) {
 			downtimeSaveCallCount: 0,
 			downtimeDuration:      time.Duration(time.Second * 8),
 			calculateDowntimeErr:  nil,
+			requestTimestamp: time.Now(),
+			lastPingTimestamp: time.Now().Add(-5 * time.Second),
 		},
 	}
 
 	for _, test := range tests {
 		t.Run(test.name, func(t *testing.T) {
 
-			timestamp := time.Now()
 			// create mock controller
 			nodeRepoMock := mocks.NodeRepository{}
 			recordRepoMock := mocks.RecordRepository{}
@@ -92,10 +103,10 @@ func TestApiController_PingHandler(t *testing.T) {
 			pingRepoMock := mocks.PingRepository{}
 			pingRepoMock.On("Save", &models.Ping{
 				NodeId:    "1",
-				Timestamp: timestamp,
+				Timestamp: test.requestTimestamp,
 			}).Return(test.pingSaveErr)
 			pingRepoMock.On("CalculateDowntime", mock.Anything, mock.Anything).Return(
-				time.Now(), test.downtimeDuration, test.calculateDowntimeErr)
+				test.lastPingTimestamp, test.downtimeDuration, test.calculateDowntimeErr)
 
 			downtimeRepoMock := mocks.DowntimeRepository{}
 			downtimeRepoMock.On("Save", mock.Anything).Return(test.downtimeSaveErr)
@@ -113,7 +124,7 @@ func TestApiController_PingHandler(t *testing.T) {
 			req, _ := http.NewRequest("POST", "/api/v1/node", bytes.NewReader(nil))
 			c := &auth.RequestContext{
 				NodeId:    "1",
-				Timestamp: timestamp,
+				Timestamp: test.requestTimestamp,
 			}
 			ctx := context.WithValue(req.Context(), auth.RequestContextKey, c)
 			req = req.WithContext(ctx)

internal/schedule/checkactive/schedule.go

@@ -34,7 +34,7 @@ func StartScheduledTask(repos *repositories.Repos) {
 func scheduledTask(repos *repositories.Repos, actions actions.Actions) {
 	log.Debug("Started task: check all active nodes")
 	activeNodes := repos.NodeRepo.GetAllActiveNodes()
-
+	var activeNodesAfterCheck []string
 	for _, node := range *activeNodes {
 
 		pingActive, err := active.CheckIfPingActive(node.ID, repos)
@@ -60,6 +60,14 @@ func scheduledTask(repos *repositories.Repos, actions actions.Actions) {
 				log.Errorf("Unable to remove node %s from active because of %v", node.ID, err)
 			}
 			log.Debugf("Node %s metrics lagging more than 10 blocks, removed node from active", node.ID)
+		} else {
+			activeNodesAfterCheck = append(activeNodesAfterCheck, node.ID)
 		}
 	}
+
+	if len(activeNodesAfterCheck) == 0 {
+		log.Debug("There is no active nodes currently")
+	} else {
+		log.Debugf("Currently active nodes: %v", activeNodesAfterCheck)
+	}
 }